Hello. Being new to DT, I’m finding it a blast to use!
I’m still setting up DT to work best with my workflow. I have realized there are pro/con’s of importing and deleting versus importing and keeping originals.
I intend to import PDF’s, Word & Pages documents, images, etc., as well as certain key email messages and threads. I simply love the idea of being able to store all of my important info in one place. I also like keeping things simple whenever possible. So, import and delete make SO much sense to me and is highly preferable. However, I’m aware that databases sometimes go South; if I import/delete, I worry about losing my valuable data.
I wanted to put this out to the members / users of this forum to learn how others are using DT with this in mind. I’m kind of stuck; I really don’t want to keep duplicates around. Thoughts? Suggestions?
I delete everything I import – but, at this point, I’m usually only importing one or a few documents at a time. No worries about losing documents if the “database goes South”, etc. The database is nothing but a specialized folder with subfolders containing your documents. (Check for yourself – in Finder, right click a .dtBase2 file and choose “Show Package Contents” – look but don’t touch )
The documents themselves are never changed by DEVONthink. DEVONthink gathers information about the documents and a concordance of words, but this is a non-invasive process.
The best security against “going South” or North or somewhere else is: have a strong backup strategy and stick to it. Backup databases when they are closed. Also, use"Verify and Repair" on your databases, especially if you have a crash on your computer for any reason. Use “Backup & Optimize” from time to time, and if you see any performance issues (a rare occurrence) use “Rebuild”. I have databases that I’ve used all day every day for years on end, and have never lost a document – except due to my own stupidity.
I started using DTPO as my academic information manager, shortly after crossing over to the Mac world. Given hard-learnt lessons over in the Windows world - I was very wary of putting all my data into a new programme.
I accordingly went with the import option, essentially duplicating 20gbs of PDFs. My “originals”, eventually found their way into my reference manager, although - it too simply ‘watches’ things like DTPO, since I don’t annotate inside of it (Bookends).
Knowing what I know now, I wish I had indexed my Bookends/academic library into DTPO. It is so solid, and has hardly given me anything to worry about over the past 3(?) years. It would also have meant that, as I add new data, I too would’ve kept things ‘simple’. Unfortunately, my OCD kicks in, and I see myself continuing along the path already laid - PDFs in both DTPO and Bookends. It gets a bit tedious at times - although the folder actions help things along - but I’d rather that than doubts about what is where [they’re ALL both there and here!]
My point is simply that, had I not started where I am now - I too would have followed the simplest option. DTPO is non-invasive. It’s easy to get things out (provided you don’t, as mentioned by korm, go do something silly) - and when they’re inside, they remain in goods hands. A solid backup regime wins every time - but a corrupt database should not give one sleepless nights, imo.
That all being said - not that I do either - but just be extra careful of syncing. When DB sharing, either to a different Mac, or through DTTG with your entire DB being “shared” on iOS - things have been known to go wrong occasionally, judging by some threads that have popped up over time. Not pointing fingers - just saying that syncing adds a certain level of complexity, which might see something go wrong, as possibly more likely.
I index everything. I’ve generally been satisfied with the results. In particular, I feel comfortable knowing that my data is backed up on the cloud (SpiderOak) and in TimeMachine. For a time, I used to import everything, and I don’t remember having any problems then. It probably just comes down to personal preference.
Syncing (DTTG on the iPad) is generally no problem for me, though it took a while to work out the best way to do it for my particular use case. The only exception would be an RTF problem a while back that took the Japanese text in any RTF I opened and corrupted it, leaving a garbled, indecipherable mess. I still have a few of those files floating around
Fortunately, nothing of great import was affected, and I had copies of most data elsewhere. The problem here was (1) I didn’t recognize the problem for a while and (2) it took me a while to go back and reconstruct stuff. You can’t just replace the corrupted files with the old backups when you have made modifications to the corrupted ones. It is quite a headache.
But, I don’t think we should be too hard on the DT developers. Something similar happened to me using Evernote. I had a massive bibliography of hundreds of sources for my dissertation corrupted. The iOS version of Evernote does not tell you when a sync problem occurs; it simply appends a version of the problem note into the existing one, so you end up (especially with multiple sync conflicts) with several versions of a note altogether in one. You don’t realize what is happening, you make modifications to the content over the course of days, weeks, or months, and then you discover that you have actually been modifying bits and pieces of multiple versions, and now none of the versions are “correct.” They all have “correct” parts, but you have to sort through them all to figure out the best one (usually, a combination of the different modifications). That occurred a few years ago and I am still working on it. Ugh.
In short: syncing is tricky, especially with iOS, so backup often, and be vigilant.
Hi FROBGOBLIN. Thanks for your thoughts. This is exactly the problem I’m concerned about. I realize nothing is 100%. But, learning from other users’ experience how widespread/often this occurs is helpful in making my decision. What I have learned from scouring the forum is that DT is EXTREMELY stable and it is very unusual for things to go awry; again, I realize nothing is 100%. The fact DT is so stable is very reassuring. Thanks for your thoughts; much appreciated!
Glad I could help! I agree with your assessment. DT is very stable. I’d also add that the developers appear to be very careful about making changes – they think things through and they generally tweak things little by little according to user feedback. The consistency over time, reliability, and helpfulness of the DT folks should also be taken into consideration when you are thinking about what to use for managing your data. It isn’t necessarily the easiest product to learn, much less master, but I think you’ll find that the time you put into it will pay off. Personally, I think it is one of the best applications I have ever used on the Mac.